Maritime rapid search and rescue method based on cooperation of multiple unmanned aerial vehicles

Abstract

The invention discloses a rapid maritime search and rescue method based on cooperation of multiple unmanned aerial vehicles, relates to the technical field of maritime rescue, and aims to improve the positioning and rescue efficiency of maritime wrecking targets. The method comprises the steps that multiple unmanned aerial vehicles are coordinated through a ground control center, search and rescue areas are dynamically divided according to the performance of the unmanned aerial vehicles, and tasks are allocated; a mixed search path combining a Z shape and a spiral shape is adopted, the credibility of a target is evaluated in combination with multi-source sensor data, and efficient search and accurate locking are achieved. Constructing a drift prediction model by using ocean current and wind speed, and dynamically adjusting a path to adapt to target drift; and finally, precisely putting the relief goods through visual identification and environment correction, and providing a fault-tolerant mechanism to ensure task continuity. The method overcomes the limitations of small coverage range and slow response of a traditional search and rescue method and the problem of insufficient dynamic environment adaptation of an existing unmanned aerial vehicle technology, significantly improves the search and rescue efficiency, accuracy and reliability, and provides an effective solution for rapid maritime rescue.

Description

Technical Field

[0001] The present invention relates to the field of maritime rescue technology, and in particular to a rapid maritime search and rescue method based on multi-UAV collaboration. Background Art

[0002] With the rapid development of the global marine economy and shipping, the frequency of maritime accidents (such as shipwrecks, people falling overboard, and aircraft disappearing) has increased annually, posing significant challenges to search and rescue efforts. Traditional maritime search and rescue efforts primarily rely on ships, helicopters, or satellite technology. Ship-based search and rescue has limited coverage, typically traveling at speeds of 10-20 knots (approximately 5-10 meters per second), making it difficult to quickly reach distant accident sites. While helicopters are relatively fast (approximately 50-80 meters per second), they are limited by high operating costs and weather conditions, making it difficult to conduct long, large-scale searches. Satellite technology can provide wide-area monitoring, but its resolution and real-time performance are insufficient, making it difficult to accurately locate floating targets or deploy relief supplies. These methods generally have low search and rescue efficiency, especially in adverse sea conditions or at night. Average positioning times can take hours or even days, missing crucial rescue time.

[0003] In recent years, drone technology has been increasingly used in maritime search and rescue due to its flexibility, low cost, and high maneuverability. Existing single-drone search and rescue systems are typically equipped with infrared cameras, radars, or optical sensors, capable of performing searches within small to medium-sized areas (e.g., 10-50 square kilometers). For example, a reported drone search and rescue solution can cover an area of ​​approximately 30 square kilometers at a speed of 20 meters per second in winds below 15 meters per second, with a positioning accuracy of 5-10 meters. However, the single-drone mode is limited by the drone's flight time (typically 30-60 minutes) and sensor coverage radius (typically 50-100 meters), making it inefficient for large-scale search and rescue missions and unable to cope with dynamic changes caused by target drift or complex sea conditions. Furthermore, single drones lack the ability to coordinate with other equipment, and if the battery runs out or a malfunction occurs, the search and rescue mission may be interrupted.

[0004] In response to the above problems, this application proposes a rapid maritime search and rescue method based on multi-UAV collaboration to solve these problems. Summary of the Invention

[0005] The purpose of the present invention is to solve the existing technical problems raised in the above background technology and provide a rapid maritime search and rescue method based on multi-UAV collaboration.

[0006] The present invention achieves the above-mentioned object through the following scheme: The solution of the present invention provides a rapid maritime search and rescue method based on multi-UAV collaboration, comprising the following steps: (1) Based on the initial information of the search and rescue mission, the improved K-means clustering algorithm is used to divide the search and rescue area into multiple sub-areas, and the remaining energy of the UAV is used to calculate the sub-areas. and battery life Dynamically allocate tasks, sub-region center distance satisfy: in, For the UAV coordinates, is the coordinate of the center of the sub-region; (2) Multiple UAVs use a hybrid search path combining "Z-shaped" and "spiral" to perform collaborative search, and the path spacing is Radius detected by sensor Identify and lock in potential targets through target credibility assessment; (3) Based on ocean current speed and wind speed Build a drift prediction model to dynamically adjust the drone path and drift displacement for: in, is the wind speed influence coefficient, is the prediction time interval; (4) Calculate the target coordinate deviation through the airborne visual recognition system and calculate the target coordinate deviation according to the wind speed. and placement height Accurately deliver relief supplies, horizontal offset satisfy: in, is the acceleration due to gravity (unit: m / s2, take 9.8); (5) When a UAV exits the mission due to insufficient energy or failure, the task allocation parameters in step (1) are re-optimized; The above steps achieve collaborative control through real-time communication between the ground control center and multiple UAVs.

[0007] Furthermore, the optimization objective function of the dynamic allocation task area division is: in, is the number of drones, is the distance weight, is the energy efficiency weight, The meaning is the same as above, and the optimization result makes each drone cover an independent sub-area.

[0008] Furthermore, the specific implementation of the hybrid search path includes: ① The spacing of the "Z"-shaped path The calculation formula is: in, is the angle between the UAV heading and the area boundary; ②The radius of the spiral path Dynamically adjust over time: in, is the attenuation coefficient, For search time.

[0009] Furthermore, the target credibility of the target credibility evaluation The calculation formula is: in, is the infrared signal strength, is the radar echo intensity, is the AIS signal matching degree, is the weight coefficient, and ,when Lock the target when

[0010] Furthermore, the path cost function of the dynamic adjustment of the drone path for: in, Same meaning as above, is the flight speed of the drone, is the adjustment parameter.

[0011] Furthermore, in the precise delivery of relief supplies, the target coordinate deviation The visual recognition system calculates in real time the placement position based on and Correction, the error is controlled within 2 meters.

[0012] Furthermore, when a UAV exits a mission due to insufficient energy or failure, the task allocation optimization objective is recalculated: in, Same meaning as above, reallocated so that the remaining drones cover all sub-areas.

[0013] Another aspect of the present invention provides a rapid maritime search and rescue system based on multi-UAV collaboration, the system comprising: a ground control center for executing the dynamic task area division and allocation; A swarm of multiple drones, each equipped with an infrared camera, radar, AIS receiver, and communication module, to perform the aforementioned search, path planning, and material delivery; Real-time communication network supports data exchange between the ground control center and the drone swarm.

[0014] Furthermore, the search coverage of the multi-UAV swarm The calculation formula is: in, For the The search area of ​​the drone, is the total search and rescue area.

[0015] Furthermore, the ground control center completes the task efficiently Monitor drone status: in, Same meaning as above, when When the pressure drops below a preset threshold, the emergency mechanism is triggered.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. The present invention provides a method for rapid maritime search and rescue based on multi-UAV collaboration, which first shows significant advantages in task allocation and search efficiency. Through an improved clustering algorithm and dynamic task allocation mechanism, the search and rescue area is divided in real time according to the performance of each UAV, avoiding waste of resources and uneven tasks, making the search and rescue process more efficient and significantly shortening the time required. At the same time, multiple UAVs adopt a hybrid search path that combines "Z-shaped" and "spiral" shapes, and cooperate with multi-source sensor data fusion to evaluate target credibility, which greatly improves the search coverage and target recognition accuracy. Compared with traditional single-path or single-sensor methods, it can lock on to the real target faster and more accurately;

[0017] 2. The present invention's improvements in path planning and material delivery further enhance the adaptability and accuracy of search and rescue operations. Taking into account the influence of ocean currents and wind speed, the system dynamically adjusts the drone's path through a drift prediction model, enabling it to adapt to the dynamic changes of targets at sea. This overcomes the limitations of existing technologies, which have fixed paths and difficulty tracking drifting targets. During the delivery of rescue materials, the system utilizes a visual recognition system and environmental factors to correct the delivery position, ensuring that the materials land accurately near the target. This significantly improves the delivery success rate compared to traditional methods, providing more timely assistance to people in the water.

[0018] 3. The present invention ensures the reliability and continuity of the system through a complete fault-tolerant and emergency mechanism. When a drone exits the mission due to insufficient energy or a malfunction, the system can quickly reallocate the mission area to avoid interruption of search and rescue due to a single point of failure. Compared with the existing technology that lacks emergency adjustment solutions, the present invention enhances stability. Combined with the efficient collaboration and intelligent design of multiple drones, it overcomes the problems of traditional search and rescue methods such as small coverage, slow response, and poor adaptability, providing a more efficient and reliable solution for rapid search and rescue at sea, greatly improving the rescue success rate and the ability to protect the safety of life and property. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the present invention or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 This is a flowchart of a rapid maritime search and rescue method based on multi-UAV collaboration of the present invention. DETAILED DESCRIPTION

[0021] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0022] Example 1: See Figure 1 As shown, this embodiment provides a rapid search and rescue method at sea based on multi-UAV collaboration, including the following steps: (1) Based on the initial information of the search and rescue mission, the improved K-means clustering algorithm is used to divide the search and rescue area into multiple sub-areas, and the remaining energy of the UAV is used to calculate the sub-areas. and battery life Dynamically allocate tasks, sub-region center distance satisfy: in, For the UAV coordinates, is the coordinate of the center of the sub-region; The optimization objective function of the dynamic allocation task area division is: in, is the number of drones, is the distance weight, is the energy efficiency weight, The meaning is the same as above, and the optimization result makes each drone cover an independent sub-area.

[0023] (2) Multiple UAVs use a hybrid search path combining "Z-shaped" and "spiral" to perform collaborative search, and the path spacing is Radius detected by sensor Identify and lock in potential targets through target credibility assessment; The specific implementation of the hybrid search path includes: ① The spacing of the "Z"-shaped path The calculation formula is: in, is the angle between the UAV heading and the area boundary; ②The radius of the spiral path Dynamically adjust over time: in, is the attenuation coefficient, For search time.

[0024] The target credibility of the target credibility evaluation The calculation formula is: in, is the infrared signal strength, is the radar echo intensity, is the AIS signal matching degree, is the weight coefficient, and ,when Lock the target when (3) Based on ocean current speed and wind speed Build a drift prediction model to dynamically adjust the drone path and drift displacement for: in, is the wind speed influence coefficient, is the prediction time interval; Dynamically adjust the path cost function of the drone path for: in, Same meaning as above, is the flight speed of the drone, is the adjustment parameter.

[0025] (4) Calculate the target coordinate deviation through the airborne visual recognition system and calculate the target coordinate deviation according to the wind speed. and placement height Accurately deliver relief supplies, horizontal offset satisfy: in, is the acceleration due to gravity (unit: m / s2, take 9.8); During the precise delivery of relief supplies, target coordinates deviated The visual recognition system calculates in real time the placement position based on and Correction, the error is controlled within 2 meters.

[0026] (5) When a UAV exits the mission due to insufficient energy or failure, the task allocation parameters in step (1) are re-optimized; When a UAV exits the mission due to insufficient energy or failure, the task allocation optimization objective is recalculated: in, Same meaning as above, reallocated so that the remaining drones cover all sub-areas.

[0027] The above steps achieve collaborative control through real-time communication between the ground control center and multiple UAVs.

[0028] Embodiment 2: This embodiment provides a rapid maritime search and rescue system based on multi-UAV collaboration, the system comprising: a ground control center for executing the dynamic task area division and allocation; A swarm of multiple drones, each equipped with an infrared camera, radar, AIS receiver, and communication module, for performing the search, path planning, and material delivery described in claim 1; Real-time communication network supports data exchange between the ground control center and the drone swarm.

[0029] Search coverage of multi-drone swarms The calculation formula is: in, For the The search area of ​​the drone, is the total search and rescue area.

[0030] Ground control center completes mission efficiency Monitor drone status: in, Same meaning as claim 5. When the pressure drops below a preset threshold, the emergency mechanism is triggered.

[0031] Example 3: This example provides a method and system for rapid maritime search and rescue based on multi-UAV collaboration, aiming to achieve efficient positioning and rescue of maritime accident targets. The following describes the implementation process of this method and system in detail with reference to specific implementation scenarios. Assume that a maritime accident occurs at 120° east longitude and 30° north latitude, the search and rescue area is 100 square kilometers, and the ocean current speed is The wind speed is 0.5 m / s The search and rescue mission is carried out by 5 drones coordinated by the ground control center. Each drone is equipped with an infrared camera, radar, AIS receiver and communication module.

[0032] At the beginning of the mission, the ground control center initializes the search and rescue area based on initial information such as the accident location, ocean current and wind speed data.

[0033] First, the 100 square kilometers search and rescue area is divided into 5 sub-areas using the improved K-means clustering algorithm to ensure that the area of ​​each sub-area is balanced and suitable for the performance of the drone. During the division process, the distance from each drone to the center of the sub-area is calculated. , the formula is ,in is the current coordinate of the drone (unit: meter), is the coordinate of the center of the sub-area (unit: meter). For example, the initial coordinates of UAV 1 are (100,100) and the center of sub-area 1 is (150,150), then rice.

[0034] Next, based on the remaining energy of each drone (Unit: Joule) and battery life (Unit: seconds) Dynamically allocate tasks, and optimize the objective function as follows: ,in , 0.001 / joule, assuming that drone 1 joule, seconds, then Joules / second, after weighted optimization with the distance term, the final allocation result is that drone 1 is responsible for sub-area 1, drone 2 is responsible for sub-area 2, and so on.

[0035] After the task allocation is completed, the five drones enter their respective sub-areas to perform collaborative searches. The search adopts a hybrid path combining "Z-shaped" and "spiral" to improve coverage efficiency.

[0036] In the initial stage, each drone flies along a "Z-shaped" path with a path spacing of , where the sensor detection radius Meters, heading angle ,but meters. If UAV 1 detects a heat source signal in sub-area 1, it switches to a "spiral" path with a spiral radius of , take the attenuation coefficient , search time seconds, then meters, gradually narrowing the search range.

[0037] To lock onto the target, each drone shares sensor data through inter-machine communication and calculates the target credibility. Assuming that target A is detected, the infrared signal strength , radar echo intensity 、AIS matching degree , weight 、 、 ,but If the threshold is 0.8, continue searching; if subsequent data updates make , then lock target A.

[0038] After the target is locked, the system enters the path planning stage. The ground control center calculates the speed of the ocean current. Meters per second and wind speed 10 m / s to build a drift prediction model and calculate the drift displacement , take the wind speed influence coefficient , prediction interval seconds, then m, UAV 1 adjusts the path according to the current position and drift displacement of the target, and the path cost function is Assume that the speed of drone 1 is m / s, rice, Meter, then m, UAV 1 flies to the predicted position of target A along the optimized path.

[0039] After reaching the target, UAV 1 performs the rescue material delivery. The visual recognition system calculates the target coordinate deviation. rice, -2 meters, launch height Meters, wind speed m / s, horizontal offset , substituting the data into rice.

[0040] Drone 1 adjusts the drop point upstream of the target to ensure that the lifebuoy landing error is less than 2 meters. After the drop, drone 1 verifies the result through the camera. If it fails, the adjacent drone 2 will take over and perform the second drop. During the search and rescue process, if drone 3 has residual energy, Insufficient (e.g., reduced to Joule) exits the mission, the system triggers the fault tolerance mechanism, and the ground control center recalculates the optimization target , the task of sub-area 3 is assigned to UAV 4 to ensure the continuity of search and rescue.

[0041] Task completion efficiency Monitoring, such as drone 1 15 / 225.33=239.6 (dimensionless). If it is lower than the threshold (such as 100), early warning and adjustment will be issued.

[0042] The entire system relies on a real-time communication network to operate, and the ground control center and the drone swarm exchange data through 5G or Ad-Hoc networks.

[0043] Search coverage , assuming that each drone searches an area Square kilometers, total area 100 square kilometers, then , and gradually increased to more than 95% through multiple rounds of searches.

[0044] Ultimately, the drone swarm located the target and completed the rescue within 2 hours, verifying the efficiency of the method.

[0045] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A rapid maritime search and rescue method based on multi-UAV collaboration is characterized by: The following steps are involved: (1) Based on the initial information of the search and rescue mission, the improved K-means clustering algorithm is used to divide the search and rescue area into multiple sub-areas, and the remaining energy of the UAV is used to calculate the sub-areas. and battery life Dynamically allocate tasks, sub-region center distance satisfy: in, For the UAV coordinates, is the coordinate of the center of the sub-region; (2) Multiple UAVs use a hybrid search path combining "Z-shaped" and "spiral" to perform collaborative search, and the path spacing is Radius detected by sensor Identify and lock in potential targets through target credibility assessment; (3) Based on ocean current speed and wind speed Build a drift prediction model to dynamically adjust the drone path and drift displacement for: in, is the wind speed influence coefficient, is the prediction time interval; (4) Calculate the target coordinate deviation through the airborne visual recognition system and calculate the target coordinate deviation according to the wind speed. and placement height Accurately deliver relief supplies, horizontal offset satisfy: in, is the acceleration due to gravity; (5) When a UAV exits the mission due to insufficient energy or failure, the task allocation parameters in step (1) are re-optimized; The above steps achieve collaborative control through real-time communication between the ground control center and multiple UAVs.

2. The rapid search and rescue method at sea based on multi-UAV collaboration according to claim 1 is characterized in that: The optimization objective function of the dynamic allocation task area division is: in, is the number of drones, is the distance weight, is the energy efficiency weight, The meaning is the same as above, and the optimization result makes each drone cover an independent sub-area.

3. The rapid search and rescue method at sea based on multi-UAV collaboration according to claim 1 is characterized in that: The specific implementation of the hybrid search path includes: ① The spacing of the "Z"-shaped path The calculation formula is: in, is the angle between the UAV heading and the area boundary; ②The radius of the "spiral" path Dynamically adjust over time: in, is the attenuation coefficient, For search time.

4. The rapid search and rescue method at sea based on multi-UAV collaboration according to claim 1 is characterized in that: The target credibility of the target credibility evaluation The calculation formula is: in, is the infrared signal strength, is the radar echo intensity, is the AIS signal matching degree, is the weight coefficient, and ,when Lock the target when 5. The rapid search and rescue method at sea based on multi-UAV collaboration according to claim 1 is characterized in that: The path cost function of the dynamic adjustment of the drone path for: in, Same meaning as above, is the flight speed of the drone, is the adjustment parameter.

6. The rapid search and rescue method at sea based on multi-UAV collaboration according to claim 1, characterized in that: During the precise delivery of the rescue supplies, the target coordinates deviate The visual recognition system calculates in real time the placement position based on and Correction, the error is controlled within 2 meters.

7. The rapid maritime search and rescue method based on multi-UAV collaboration according to claim 1 is characterized in that: When a UAV exits the mission due to insufficient energy or failure, the task allocation optimization objective is recalculated: in, Same meaning as above, reallocated so that the remaining drones cover all sub-areas.

8. A rapid maritime search and rescue system based on multi-UAV collaboration, characterized in that: include: A ground control center, configured to execute the dynamic task area division and allocation; A swarm of multiple drones, each equipped with an infrared camera, radar, AIS receiver, and communication module, for performing the search, path planning, and material delivery described in claim 1; Real-time communication network supports data exchange between the ground control center and the drone swarm.

9. The rapid maritime search and rescue system based on multi-UAV collaboration according to claim 8, characterized in that: Search coverage of the multi-UAV swarm The calculation formula is: in, For the The search area of ​​the drone, is the total search and rescue area.

10. The rapid maritime search and rescue system based on multi-UAV collaboration according to claim 8, characterized in that: The ground control center completes the task efficiently Monitor drone status: in, Same meaning as claim 5. When the pressure drops below a preset threshold, the emergency mechanism is triggered.

Images